Prenatal exposure to alcohol can have lasting effects on functioning in attentional, cognitive, social, and behavioral domains, and is a major cause of mental retardation. Infants born to African-American and/or low SES women appear to be at increased risk of adverse effects. Screening, brief intervention, and referral for treatment (SBIRT) approaches are a promising option for addressing this significant need: many studies have suggested that these approaches can be successfully used in primary care settings to identify and reduce alcohol use, and their brief proactive nature gives them substantial potential for reaching otherwise unidentified at-risk women. Unfortunately, growing evidence suggests that the promise of SBIRT approaches may be limited by the amount of time, training, expertise, and commitment they require. Computer-delivered SBIRT approaches may address this limitation by providing consistent screening and evidence-based brief interventions, at low cost, without requiring substantial investments of time or energy from medical staff. However, several Stage I steps are necessary before moving to a Stage II clinical trial. This R34 application will therefore lay the groundwork for a fully powered clinical trial of a computer-delivered SBIRT for alcohol use during pregnancy. It will do so through the conduct of five key preliminary studies, including: (1) evaluation of the utility of handheld mobile devices and an anonymous self-interview format in screening for at-risk drinking among patients attending a prenatal clinic; (2) modification of an existing computer-delivered motivational intervention for alcohol use during pregnancy, to a priori standards of acceptability (to experts as well as representative pregnant women); (3) development of an evidence-based tailored messaging supplement to the single-session brief intervention; (4) examining the validity of, and cut scores for, the biomarker Ethyl Glucoronide (EtG) in pregnant women; and (5) collecting data on the acceptability, feasibility, and estimated effect size of the modified computer-delivered intervention through an N = 50 Phase I randomized clinical trial. Participants in this trial will be a diverse sample of women at-risk for alcohol use during pregnancy, the majority of whom will be African-American and/or low SES. These key preparatory steps will greatly facilitate the subsequent development of an R01 application to conduct a Stage II clinical trial for alcohol use during pregnancy. These steps will also provide important preliminary data on (a) a novel method for risk factor screening in primary care; (b) the potential utility of EtG as a biomarker for alcohol use during pregnancy and in the perinatal period; and (c) the effect size estimate for a fully computer-delivered, combined brief interactive/tailored messaging intervention requiring only a single contact. If successful, this line of research could lead to a highly cost-effective, high-reach intervention for alcohol use during pregnancy; these reductions in alcohol use could in turn have a meaningful population impact on Fetal Alcohol Spectrum Disorders.